Depends on what you want to do with the previous responses. eg. if piping survey 1 responses back into survey 2, could perhaps try:



* Basing the "survey 2 distribution list" off the survey 1 data / csv file (ie. leaving in individual records with contact details + key survey 1 responses)



* Then, build in embedded variables into survey 2, that are based off "survey 1 response" data fields



* Then pipe these into the latest survey

---

Thanks Jenny, my concern with this approach (which I think is the only technically viable one in Q) is that, as the survey library of possible carry forward embedded fields grows across multiple surveys, managing the embedded fields that need to be created in survey flow at the start of every survey that might use carry forward could be unmanageable. It may be that I am just trying to be too ambitious in using past responses to contextualise future surveys to the same contacts.

---